Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Deep learning in acute ischemic stroke imaging: a systematic review of CT- and MRI-based segmentation, triage, and prognostic modeling.

Neuroradiology·2026

Same author

Robust artificial intelligence frameworks for lung cancer subtyping and malignancy detection on thoracic CT.

Scientific reports·2026

Same author

DiSCNet: Directional Split Convolution for compute-efficient brain tumor diagnosis.

Computational biology and chemistry·2026

Same author

Behavioral economics-oriented energy storage investment analysis: A holistic decision support model with advanced fuzzy techniques.

Journal of environmental management·2026

Same author

ThermoMicrowave-sonication improves the stability and digestive bioaccessibility of phenolic compounds in parsley juice.

Food chemistry: X·2026

Same author

A systematic review of machine learning in heart disease prediction.

Turkish journal of biology = Turk biyoloji dergisi·2025

Search research articles

相关实验视频

Updated: Jun 14, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

NeXtBrain:结合本地和全球特征学习来进行脑瘤分类.

Ishak Pacal¹, Ozan Akhan², Rumeysa Tuna Deveci³

¹Department of Computer Engineering, Faculty of Engineering, Igdir University, 76000 Igdir, Turkey; Department of Electronics and Information Technologies, Faculty of Architecture and Engineering, Nakhchivan State University, AZ 7012, Nakhchivan, Azerbaijan.

|June 9, 2025

概括

NeXtBrain是一种全新的混合深度学习架构,在保持计算效率的同时,在脑瘤分类方面实现了高准确性. 该模型有效地捕捉了本地和全球瘤特征,优于现有的最先进的方法.

关键词:

脑瘤检测脑瘤检测脑瘤检测深度学习是一种深度学习.卫生健康卫生健康卫生健康医学成像医学成像视觉变压器视觉变压器

更多相关视频

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Published on: August 16, 2020

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

相关实验视频

Last Updated: Jun 14, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Published on: August 16, 2020

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

科学领域:

医学图像分析医学图像分析
人工智能的人工智能
计算病理学计算病理学

背景情况:

准确的脑瘤诊断对于治疗和患者的结果至关重要.
深度学习模型在医学图像分析中努力平衡准确性,概括性和效率.
在计算上捕获本地和全球瘤特征是具有挑战性的.

研究的目的:

介绍NeXtBrain,一个新的混合深度学习架构用于脑瘤分类.
克服现有模型在捕捉各种瘤特征和计算成本方面的局限性.
为了实现大脑瘤诊断的高准确性,稳定性和效率.

主要方法:

开发了NeXtBrain,这是一个混合架构,包括NeXt卷积块 (NCB) 和NeXt变压器块 (NTB).
对于局部特征提取,NCB使用多头卷积注意和基于SwiGLU的MLP.
NTB集成了自我注意力,卷积注意力和SwiGLU MLP,用于全球上下文建模.

主要成果:

在Figshare数据集上,NeXtBrain实现了99.78%的准确性和99.77%的F1得分.
在Kaggle数据集上,NeXtBrain获得了99.78%的准确性和99.81%的F1分数.
性能优于包括ViT,CNN和混合方法在内的17个最先进的模型,参数显著减少 (23.91M) 和计算成本降低 (10.32 GFLOPs).

结论:

NeXtBrain为脑瘤分类提供了一个计算效率高和高度准确的解决方案.
混合架构有效地捕捉了细粒度的本地和远程的全球瘤特征.
NeXtBrain代表了医疗图像分析深度学习的重大进步,使得更好的临床决策成为可能.